Tailoring magnetism of multifunctional MnxGa films with giant perpendicular anisotropy
L. J. Zhu, D. Pan, S. H. Nie, J. Lu, J. H. Zhaoa)
L.J.Zhu et al.,arXiv:1210.2062(2012)
State Key Laboratory of Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of
Sciences, P. O. Box 912, Beijing 100083, China
Abstract: We report wide-range composition and annealing effects on magnetic properties of MnxGa films
grown on GaAs (001) by molecular-beam epitaxy. We obtained single-crystalline MnxGa films in a surprisingly
wide composition range from x=0.76 to 2.6. We show that the magnetism could be effectively tailored by
adjusting composition and annealing. Especially, when 0.76≤x≤1.75, MnxGa films simultaneously show
magnetization from 130 to 450 emu/cc, perpendicular anisotropy among 8.6 to 21 Merg/cc, intrinsic coercivity
from 4.38 to 20.1 kOe, normal coercivity up to 3.6 kOe, energy product up to 3.4 MGOe and thermal-stability
up to at least 350 oC in contact with GaAs.
Keywords: High coercivity materials, Perpendicular magnetic anisotropy, Spintronics, Molecular-beam epitaxy.
PACS numbers: 75.50.Vv, 75.30.Gw, 85.75.-d, 81.15.Hi
a)
Author to whom correspondence should be addressed;
Electronic mail: jhzhao@red.semi.ac.cn
metals (e.g. Cr and Pt).15-19 However, there is few
study about D022-MnxGa films on semiconductors.
Meanwhile, although the growth of L10-MnxGa has
been studied on many semiconductors including GaN,
ScN, GaSb, Si and GaAs, only a few of those films
on ScN (x=1.5) and GaAs (x=1.2-1.5) demonstrated
perpendicular easy axis with small magnetization and
coercivity well below 230 emu/cc and 8.8 kOe at
room temperatures, respectively.8,10,18,20-22 Also, there
has been a shortage of the detailed perpendicular
anisotropy of these L10-MnGa films. We recently
presented L10-Mn1.5Ga films on GaAs (001) with Ms,
Hc, Ku and (BH)max up to 270.5 emu/cc, 42.8 kOe,
21.7 Merg/cc and 2.6 MGOe, respectively.2 However,
Ms and (BH)max of these films were still smaller than
theory-predicted values.12-15 It is therefore highly
desirable to investigate some effective ways to
control and optimize the magnetism for various
potential applications, such as by tuning composition
and post-growth annealing.
There have been some reports on the
composition-dependence of magnetic properties of
MnxGa polycrystalline bulks,14,23 in which it is
difficult to estimate the magnetic anisotropy and
intrinsic magnetism. Also, all growth of MnxGa films
ever performed was limited to very narrow
composition range belonging to either L10 or D022
phases.8,10,15-18,20-22 Recently, (001)-orientated MnxGa
with x=1.2-3 and (111)-orientated MnxGa with
x=0.96-2.03 have been studied on Cr-MgO and GaN,
respectively.19,22 However, there have not been
systematic studies on wide-range composition effects
on magnetic properties of MnxGa single-crystalline
films with (001)-orientation, especially those on
semiconductors. Moreover, an effective tailoring of
magnetic properties by post-growth annealing could
be expected since they are very sensitive to growth
temperature.2 Still, there are few detailed studies
devoted to post-growth annealing effects on magnetic
properties in MnxGa films.
Magnetic materials simultaneously with high
perpendicular anisotropy (Ku), coercivity (Hc) and
energy products ((BH)max) have great application
potential in ultrahigh-density perpendicular magnetic
recording, permanent magnets and spintronic devices
including spin-transfer-torque (STT) magnetic
random access memory (MRAM) and oscillators,
and magnetoresistance sensors.1-7 On the other hand,
magnetic materials compatible with semiconductors
in epitaxy allow for direct integration of magnetic
devices with electronics or optical underlying
Moreover,
ferromagnetic
circuitry.8
metal/semiconductor heterostructures have great
potential in novel applications such as spin-based
light emitting diodes and field effect transistors.8-11
Thus, it is desirable to develop novel kind of
perpendicular magnetized materials epitaxied on
semiconductors and further tailor their magnetic
properties to meet corresponding demands for
various functional applications.
In the past two decades, noble-metal-free and
rare-earth-free MnxGa films with L10 (x<2) or D022
(2<x<3) structures (Fig. 1(a)) have attracted
increasing attention for their theory-predicted
magnetic properties remarkably desirable for
applications in ultrahigh-density magnetic recording,
high-performance spintronic devices and economical
permanent magnets. L10-MnGa (D022-Mn3Ga) alloys
with (001)-orientation are theoretically expected to
have Ku of 26 (20) Merg/cc, magnetization (Ms) of
845 (305) emu/cc, (BH)max of 28.2 (3.67) MGOe,
spin polarization of 71% (88%) at the Fermi level
and Gilbert damping constant of 0.0003 (0.001),
respectively.12-15 To realize practical applications of
this kind of material in magnetic devices compatible
with semiconductor circuitry, it is essential to explore
the growth of single-crystalline L10 and D022 MnxGa
films with (001)-orientation on semiconductors. So
far, many attempts have been made on the growth of
MnxGa films on insulators (e.g. MgO and Al2O3) and
1
L.J.Zhu et al.,arXiv:1210.2062(2012)
FIG. 1 (a) Lattice units of L10-MnGa (left) and D022-Mn3Ga (right), arrows stand for magnetic moment directions. Typical
(b) XRR curves, (c) High-sensitivity XPS, (d) Synchrotron XRD patterns, (e) FWHM of MnxGa fundamental peaks, (f)
Isup/Ifun and (g) c of MnxGa films. The blue arrows in (d) point to the (002), (003) and (004) peaks of MnGa4 phase appeared
in the Mn0.55Ga and Mn0.74Ga. The c value for MnxGa bulks in ref. 14 (red ●) and ref. 23 (green ●) and films grown on GaN
in ref. 21 (wine ●) are also plotted in (g), for comparison with films grown on GaAs (001) (wine O).
In this letter, we present synthesis of (001)orientated single-crystalline MnxGa films with giant
perpendicular anisotropy epitaxied on GaAs (001) in
a surprisingly wide composition from x=0.76 to 2.6,
and effective tailoring of the magnetic properties by
controlling composition and annealing. These results
would be much helpful for understanding this kind of
materials and approaching functional applications in
magnetic recording, spintronic devices and
permanent magnets.
A series of MnxGa films with different
Mn/Ga atom ratio x were deposited at 250 oC on 150
nm-GaAs-buffered semi-insulating GaAs (001)
substrates by molecular-beam epitaxy under base
pressure less than 1×10-9 mbar.2 After cooling down
to room temperature, each film was capped with a
1.5 nm-MgO layer to prevent oxidation. The
composition was designed by carefully controlling
the Mn and Ga fluxes during growth and later
verified by high-sensitivity x-ray photoelectron
spectroscopy
(XPS)
measurements
(Thermo
Scientific ESCALAB 250Xi) with Al Kα source and
relative atomic sensitivity factors of 13.91 (1.085) for
Mn 2p (Ga 3d). The thicknesses and crystalline
structures of the films were respectively determined
by synchrotron radiation x-ray reflection (XRR) and
x-ray diffraction (XRD) with wave length of 1.54 Å.
The magnetization measurements were performed by
superconducting quantum interference devices
(SQUID) up to 5 T and physical properties
measurement system (PPMS) up to 14 T. The
annealing studies were performed at temperatures
from 100 to 500 oC under vacuum below 110-8 mbar.
Figure 1(b) shows typical XRR curves with the
strong oscillations, suggesting sharp interfaces and
good homogeneity of these samples. From XRR
curves, the thicknesses were also determined to range
from 35 to 50 nm. Figure 1(c) shows examples of
high-sensitivity XPS of Mn 2p and Ga 3d in the
MnxGa films. From XPS results, x was estimated to
be 0.55, 0.74, 0.76, 0.93, 0.94, 0.97, 1.07, 1.13, 1.29,
1.42, 1.75 and 2.60, respectively. The asymmetry of
these peaks should be attributed to the significant
shielding effect to core levels by the high statedensity at the Fermi level due to the alloy nature of
these films. Furthermore, no considerable peak shifts
are observable for both the Mn 2p and Ga 3d
spectrums in all these films because of the protection
of MgO cap layers from oxidation.
Figure 1(d) shows examples of XRD θ-2θ patterns of
MnxGa films with different x. For x ≥ 0.76, only
sharp (001) ((002)) superlattice peaks and (002)
((004)) fundamental peaks of L10 (D022) MnxGa
films can be observed in the range from 20o to 70o
besides the peaks of GaAs substrates, indicating that
these are all (001)-textured single-crystalline films.
For film with x=0.74, three small peaks irrelative to
L10 (or D022) appear, which become dominating for
film with x= 0.55. The new phase could be best fitted
by cubic MnGa4 24 with enlarged c axis of 3.91 Å,
while Mn2Ga5, MnGa6, Mn3Ga5 and Mn5Ga7 phase24,
25
are not the case here. Noticeably, the (001)orientated D022-Mn2.6Ga film is of very poor
crystalline quality and chemical ordering, indicating
that it is difficult to crystallize the metastable D022MnxGa with high quality on GaAs, even when the
lattice mismatch is very small (~2.2%). The full
width at half maximum (FWHM) of fundamental
2
L.J.Zhu et al.,arXiv:1210.2062(2012)
composition range are magnetically phase-pure with
perpendicular easy axis. Figure 2(b) shows typically
both perpendicular and in-plane hysteresis loops of
MnxGa films. The perpendicular M-H curves show
square-like shapes, whereas in-plane curves exhibit
almost anhysteretic loops and high saturation field
even exceeding 14 Tesla. The strong difference
between the perpendicular curves and in-plane curves
reveals giant perpendicular anisotropy in these films.
In consistence with XRD results, MnxGa films with
x=0.55 and 0.74 exhibit weak perpendicular
anisotropy with very close Mr-T and M-H curves in
perpendicular and in-plane directions due to the
coexisting of MnGa4 phase and L10-phase.
Figures 3(a)-(c) display the x-dependent
magnetic properties including squareness (Mr/Ms), Ms
and Hc determined from the 300 K perpendicular MH curves of these 250 oC-grown films. The MnxGa
films with x=0.97-1.75 exhibit high Mr/Ms exceeding
0.90; but those with x deviated from this range show
dramatically decreased Mr/Ms, the tendency of which
is just opposite to that of c in Fig. 1(g), indicating
that Mr/Ms in this kind of material could be strongly
affected by strains. As shown in Fig. 3(b), Ms
decreases dramatically from 450 to 52 emu/cc with x
increases from 0.76 to 2.60, which should be
attributed partly to the increase of antiferromagnetic
coupling between Mn atoms at different sites,12
partly to the increase of strains induced by short c
axis.2,22 The maximum Ms of 445 emu/cc in Mn0.76Ga,
is still below the calculated value of 845 emu/cc for
stoichiometric L10-MnGa17 and experimental value
of 600 emu/cc in 400-450 oC-annealed Mn1.17Ga
grown on MgO 19 probably due to the low growth
temperature. Hc climbs up from 4.38 to 20.1 kOe as x
increases from 0.76 to 1.75, and then drops to 1.1
kOe at x=2.60. This change tendency is consistent
well with that of Ku estimated in Fig. 3(d) following
the method used in ref. 2. Ku exhibits large values
from 8.6 to 21.0 Merg/cc in L10 range, while quickly
degrades to 0.02 Merg/cc in D022-Mn2.6Ga because of
increased disorder.
As ever discussed,2,14 rare-earth-free and noblemetal-free MnxGa alloys are considerable for
economical permanent magnet application. Since
intrinsic coercivity Hc determined from M-H curves,
normal coercivity HcB determined from B-H curves
and magnetic energy products (BH)max are three key
features of quality of permanent magnets, we further
calculated HcB and (BH)max in Fig. 3(e) and (f). With
increasing x in the phase-pure range, both HcB and
(BH)max
decrease
nearly
monotonically,
corresponding to the decrease of Ms. The film with x
= 1.07 exhibit the highest HcB and (BH)max of 3.6 kOe
and 3.4 MGOe, respectively. The maximum of 3.4
MGOe is larger than that recent reported in Mn1.5Ga
film (2.6 MGOe)2 and ferrite magnets (3 MGOe),3
which makes L10-MnxGa alloys with low Mn
compositions still promising to be developed for
cost-effective and high-performance permanent
magnets applications.
FIG. 2 (a) Temperature dependence of remanent
magnetization (Mr) and (b) Hysteresis loops of Mn1.4Ga
film measured at 300 K.
FIG. 3 (a) Mr/Ms, (b) Ms, (c) Hc, (d) Ku, (f) HcB and (g)
(BH)max of MnxGa films on GaAs (001) plotted as a
function of x.
peaks, the integrated intensity ratio of superlattice
and fundamental peaks Isup/Ifun (proportional to
ordering parameters) and out-of-plane lattice
parameters c (c/2) of L10 (D022) MnxGa determined
from the XRD results are summarized as a function
of x in Figs. 1(c), (d) and (e), respectively. With
exception of Mn2.6Ga which is of poor quality and
ordering, MnxGa films with x ≥ 0.76 show
comparable FWHM and Isup/Ifun, implying similar
crystalline quality and long range ordering degree.
These films exhibit much short c-axis than bulks with
same compositions probably due to tensile strains
from epitaxy with GaAs. c first drops dramatically,
consistent with the trend in bulks14,23 and film on
GaN21; while it slowly comes back toward bulk
values when x further increases. This non-monotonic
trend probably relates to both tensile strains from
epitaxy on GaAs and complicated compositiondependence of high level of dislocation induced by
strain relaxation (see Supplemental Information).
Figure 2(a) shows an example of temperature
dependence of perpendicular remanent magnetization
(Mr) of MnxGa film deposited on GaAs (001) with
Curie temperature of 630 K. In contrast, the in-plane
Mr-T curves shows nearly zero Mr values in the
whole temperature range (not shown here). The
similar feature holds for all the films with x=0.762.60, revealing that all these films over the large
3
L.J.Zhu et al.,arXiv:1210.2062(2012)
FIG. 4 (a) XRD of 10 min-annealed Mn0.76Ga films at different temperature. (b) Perpendicular hysteresis loops of Mn0.76Ga
films annealed at 450 oC for different time. Annealing temperature Ta (blue O, 10 min) and time (red ●, 450 oC) dependence
of (c) Mr/Ms, (d) Ms, (e) Hc, (f) HcB, (g) (BH)max and (h) Ku of Mn0.76Ga films.
least 350 oC in respect to both structure and magnetic
properties could still satisfy the demand of
integration
with
metal-oxide-semiconductor
transistors.
It should be motioned that further annealing at
450 oC for a longer time would significantly affect
the magnetic behaviors of Mn0.67Ga films as show in
Fig. 4(b). We should attribute this change mainly to
both the structural improvement of Mn2Ga layer and
increasing Mn2As formed at the interface with
substrates. As summarized in Figs. 4 (c)-(h), with
annealing time increase from 10 to 60 min, Hc drops
monotonically from 4 to 1 kOe; Ms increases from
445 to 700 emu/cc; the highest Mr/Ms≈1 and highest
Ku were obtained after annealed at 450 oC for 20-30
min. Therefore, for typical L10-Mn0.76Ga film,
annealing at 450 oC for 20-30 min would be helpful
to applications like magnetic recording and STT
devices which favor high Mr/Ms, Ms, Ku and moderate
Hc at the same time. From the viewpoint of
permanent magnet applications, the best should be
annealing at 450 oC for 10-20 min since the largest
HcB and (BH)max of 4.4 MGOe were obtained in this
range.
In summary, both the wide-range composition
and detailed annealing effects on the structure and
magnetic properties of perpendicularly magnetized
MnxGa epitaxial films on GaAs (001) have been
systematically investigated. We show that singlecrystalline MnxGa films with L10 or D022 ordering
could be crystallized at 250 oC on GaAs along (001)
direction in a surprisingly wide composition range
from x=0.76 to 2.6. Noticeably, L10-ordered MnxGa
Taking into consideration of all the magnetic
properties in Fig. 3, low-Mn-composition L10 MnxGa
films are more advantageous than those D022 film to
be developed for applications in magnetic recording
with areal density over 10 Tb/inch2, highperformance nanoscale spintronic devices like
MRAM bits with size down to several nm and
economical permanent magnet applications, since
they simultaneously exhibit high Mr/Ms, Ms, Ku, Hc,
HcB and (BH)max.
In order to investigate the thermal-dynamical
properties and further tailor the magnetic properties
of MnxGa films, especially the Ga-rich singlecrystalline films, we performed post-growth
annealing experiments taking L10-Mn0.76Ga film as
an example, which also exhibits the largest Ms among
all the as-grown films with different x (Fig. 2).
Figure 4(a) shows XRD patterns of as-grown and 10
min-annealed Mn0.76Ga films, from which we found
the Mn0.76Ga film itself is thermal-dynamically stable
up to 450 oC, and could keep stable up to at least 350
o
C in contact with GaAs. This result roughly agrees
with previous report that MnxGa with x<1.5 should
keep stable on GaAs below 400 oC.26 Interestingly,
very weak Mn2As (003) peak as a result of interfacial
reaction is observable by high-sensitivity synchrotron
XRD measurement, which may be not detectable by
common XRD with Cu Kα radiation. 26 As
summarized in Figs. 4(c)-(h), all the magnetic
properties only exhibit very weak dependence on Ta
in the case of 10-min annealing even when a small
mount of Mn2As formed at the interface after 450 oCannealing. Fortunately, the stable temperature of at
4
L.J.Zhu et al.,arXiv:1210.2062(2012)
8
films showed high Ms, Mr/Ms, Ku, Hc, HcB and
(BH)max, which make this kind of materials favorable
for applications in ultrahigh-density perpendicular
magnetic recording, permanent magnets, highperformance spintronic devices. In contrast, D022Mn2.6Ga films exhibited lower perpendicular
anisotropy and weaker magnetism due to poor
quality and high disordering. Annealing studies
revealed thermal-stability of MnxGa up to at least
350 oC in contact with GaAs and effective tailoring
of magnetic properties by controlling the formation
of Mn2As at the interface through prolonged
annealing at 450 oC. These results would be much
helpful for understanding this kind of material and
approaching the functional applications in magnetic
recording, spintronic devices and permanent magnets.
M. Tanaka, J. P. Harbison, J. DeBoeck, T. Sands, B.
Philips, T. L. Cheeks, and V. G. Keramidas, Appl.
Phys. Lett. 62, 1565 (1993).
9
C. Adelmann, J. L. Hilton, B. D. Schultz, S.
McKernan, C. J. Palmstrøm, X. Lou, H. S. Chiang,
and P. A. Crowell, Appl. Phys. Lett. 89, 112511
(2006).
10
E Lu, D. C. Ingram, A. R. Smith, J. W. Knepper,
and F.Y. Yang, Phys. Rev. Lett. 97, 146101 (2006).
11
H. C. Koo, J. H. Kwon, J. Eom, J. Chang, S. H.
Han, and M. Johnson, Science 325, 1515 (2009).
12
A. Sakuma, J. Magn. Magn. Mater. 187, 105 (1998).
13
Z. Yang, J. Li, D. Wang, K. Zhang, and X. Xie, J.
Magn. Magn. Mater. 182, 369 (1998).
14
J. Winterlik, B. Balke, G. H. Fecher, C. Felser, M.
C. M. Alves, F.Bernardi, and J. Morais, Phys. Rev. B
77, 054406 (2008).
15
S. Mizukami, F. Wu, A. Sakuma, J. Walowski, D.
Watanabe, T. Kubota, X. Zhang, H. Naganuma, M.
Oogane, Y. Ando, and T. Miyazaki, Phys. Rev. Lett.
106, 117201 (2011).
16
F. Wu, S. Mizukami, D. Watanable, H. Naganuma,
M. Oogane, and Y. Ando, Appl. Phys. Lett. 94,
122503 (2009).
17
H. Kurt, K. Rode, M. Venkatesan, P. Stamenov,
and J. M. D. Coey, Phys. Status. Solidi B 248, 2338
(2011); Phys. Rev. B 83, 020405(R) (2011).
18
W. Feng, D. V. Thiet, D. D. Dung, Y.Shin, and S.
Chob, J. Appl. Phys. 108, 113903 (2010).
19
S. Mizukami, T. Kubota, F. Wu, X. Zhang, T.
Miyazaki, H. Naganuma, M. Oogane, A. Sakuma,
and Y. Ando, Phys. Rev. B 85, 014416 (2012).
20
K. M. Krishnan, Appl. Phys. Lett. 61, 2365 (1992).
21
A. Bedoya-Pinto, C. Zube, J. Malindretos, A.
Urban, and A. Rizzi, Phys. Rev. B 84, 104424 (2011).
22
K. Wang, A. Chinchore, W. Lin, D. C. Ingram, A.
R. Smith, A. J. Hauser, and F. Yang, J. Cry. Growth
311, 2265 (2009).
23
T. A. Bither and W. H. Cloud, J. Appl. Phys. 36,
1501 (1965).
24
H.G. Meissner and K. Schubert, Z. Metallkd. 56,
523 (1965).
25
J. S. Wu and K. H. Kuo, Metall. Mater. Trans. A
28A, 729 (1997).
26
J. L. Hilton, B. D. Schultz, S. McKernan, and C. J.
Palmstrøm, Appl. Phys. Lett. 84, 3145 (2004).
We gratefully thank G. Q. Pan at the U7B
beamline of National Synchrotron Radiation
Laboratory (NSRL) and W. Wen at the U7B
beamline of Shanghai Synchrotron Radiation Facility
(SSRF) in China for their help with XRD and XRR
measurements. This work was supported by the
NSFC under Grant Nos. 60836002 and 10920101071.
References
C. Chappert, A. Fert, and F. N. V. Dau, Nature
Mater, 6, 813 (2007).
2
L. J. Zhu, S. H. Nie, K. K. Meng, D. Pan, J. H. Zhao,
and H. Z. Zheng, Adv. Mater. 24, 4547 (2012).
3
O. Gutfleisch, M. A. Willard, E. Brück, C. H. Chen,
S. G. Sankar, and J. P. Liu, Adv. Mater. 23, 821
(2011).
4
S. Ikeda, K. Miura, H. Yamamoto, K. Mizunuma, H.
Gan, M. Endo, S. Kanai, J. Hayakawa, F. Matsukura,
and H. Ohno, Nature Mater. 9, 721(2010).
5
D. Houssameddine, U. Ebels, B. Delaët, B.
Rodmacq, I. Firastrau, F. Ponthenier, M. Brunet, C.
Thirion, J. P. Michel, L. Prejbeanu-Buda, M. C.
Cyrille, O. Redon, and B. Dieny, Nature Mater. 6,
447 (2007).
6
F. B. Mancoff, J. H. Dunn, B. M. Clemens, and R. L.
White, Appl. Phys. Lett. 77, 1879 (2000).
7
T. J. Nummy, S. P Bennett, T. Cardinal, and D.
Heiman, Appl. Phys. Lett. 99, 252506 (2011).
1
5